The detection of specific nucleic acid sequences is immensely useful in molecular medicine. The possibilities for useful detection and quantitation of specific genes and gene products are nearly endless. Genotyping methods are of interest for prenatal diagnosis; as well as detecting changes in genotype associated with disease, for example during oncogenesis. Genotyping methods also find use in pharmacogenomics, to determine an individual's profile for drug metabolism, including the likelihood of adverse reactions and responsiveness to treatment. Other important areas of research include analysis of mRNA for expression, alternative splicing and SNP variation. In addition to analysis of expression, and of sequence polymorphisms, there is significant interest in simply determining whether a target sequence is present in a sample, for example in the detection and identification of microbial species in clinical and environmental samples.
Fast, simple and accurate methods of detecting and analyzing the presence or absence of nucleic acids, which may differ by as little as one nucleotide from others, are of great interest. In some cases, the nucleic acids may be present in minute quantities or concentrations, which underscores the need for high sensitivity as well. Many methods of detecting the presence of nucleic acid sequences are known in the art, including Northern and Southern blots, microarray hybridization, and the like. These methods have typically relied on hybridization kinetics between the target and probe species, coupled with varying temperature and ionicity to provide specificity. However, there are some significant drawbacks to these methods in terms of specificity and sensitivity.
A number of laboratories have investigated the use of nonenzymatic fluorescence based approaches for RNA or DNA detection, relying on the formation of bonds, hybridization of fluorescent oligonucleotides, or changes in secondary structure to detect genetic sequences. For example, see Xu and Kool (1997) Tetrahedron Lett. 38:5595-5598; Paris et al. (1998) N.A.R. 26:3789-3793; Okamoto et al. (2003) JACS 125:9296-9297; Tyagi and Kramer (1996) Nat. Biotech. 14:303-308; Kuhn et al. (2001) Antisense Nucleic Acid Drug Dev. 11:265-270; and International Patent application WO 2004/010101. Also of interest are U.S. Pat. No. 5,571,903 (Gryaznov); and U.S. Pat. No. 4,958,013 (Letsinger).
Self-ligation reactions have been developed for sequence detection, where the chemistry for joining two short oligonucleotide probes is incorporated into the ends of the probe molecules themselves. Such self-ligation reactions can be highly selective for single nucleotide differences in the target molecule. See, for example, Xu and Kool (2000) JACS 122:9040-9041; Xu et al., (2001) Nat. Biotech. 19:148-152; Ficht et al. (2004) JACS 126:9970-9981; and Gryaznov and Letsinger (1993) JACS 115:3808-3809. In a recent advance, the chemistry for ligation was activated by a group that acted both as leaving group and as fluorescence quencher, thus enabling the probes to become fluorescent in the presence of a complementary target (Sando and Kool (2002) J. Am. Chem. Soc. 124 (10): 2096-2097).
Previously described self-ligation reactions have also had some limitations, however. The rate of reaction can be significantly slower than enzymatic ligations. In addition, prior art methods have placed the DNA end activation directly on a terminal T nucleoside, which would require the synthesis of four different modified nucleosides for application to all sequences. Finally, ligation reactions typically yield only one signal per target molecule, because the ligation product is stably bound to the target, and therefore does not readily dissociate.
While nonenzymatic approaches can offer advantages in cost, simplicity, and in vivo utility, they have not thus far provided for appreciable amplification of signals, which is useful in detecting target sequences present at low concentrations or in low numbers. Structures of suitable probes that address this issue, and provide for simplified probe preparation, are described herein.
Other Publications
Strategies for using pairs of modified oligonucleotides to generate amplified products or signals have been described. Ma and Taylor, Proc. Natl. Acad. Sci. USA 2000, 97, 11159-11163; Ma and Taylor, Bioorg. Med. Chem. 2001, 9, 2501-2510; and Brunner et al., J. Am. Chem. Soc. 2003, 125, 12410-12411 have described the combination of a hydrolysis catalyst on one oligonucleotide with a leaving group (in the form of an ester) on the other, resulting in the release of multiple leaving groups for each targeted complementary strand of DNA. Those approaches have generated ca 3-35 turnovers. The former has reported the generation of fluorescence signals, albeit without the demonstration of turnover (Ma and Taylor, Bioconjugate Chem. 2003, 14, 679-683). None of these approaches rely on ligation.
Ligations of amino-conjugated oligonucleotides have been investigated by Luther et al., Nature 1998, 396, 245-248 and by Zhan and Lynn, J. Am. Chem. Soc. 1997, 119, 12420-12421. The former approach requires denaturation cycles for turnover. The latter strategy isothermally generates as much as >50 turnovers in ligation, but it requires a separate reagent (borohydride), and it is not clear how the approach could generate easily detectable signals, such as those provided by the present invention.
Ficht et al., supra. developed peptide nucleic acid (PNA) probes that ligate by native chemical ligation; such probes have not been demonstrated to undergo turnover, nor do they generate fluorescent signals. RNA-detecting ribozymes are well documented to undergo turnover (Wang and Sen, J. Mol. Biol. 2001, 310, 723-734; Komatsu et al., Biochemistry 2002, 41, 9090-9098); however, a recent example of a DNAzyme designed to generate fluorescence signals in detection of an RNA documented only 4 signals per target (Sando et al., J. Am. Chem. Soc. 2003, 125, 15720-15721).